We propose to conduct a sigmoidoscopy-based case-control study of nutritional factors in the etiology of adenomatous polyps of the large bowel. There will be 530 cases identified from sigmoidoscopy as having a first time diagnosis of at least one adenomatous polyp. Five hundred and thirty controls, matched by age, gender, and calendar date of exam, will be selected from subjects who were found to be free of polyps at the sigmoidoscopy. The study will be restricted to asymptomatic subjects with no history of polyps. Data will be obtained from a self-administered dietary questionnaire (Walter Willetts'), nondietary risk factor questionnaire (in-person interview), blood sample, and, from cases only, pathology specimens. Dietary factors of particular interest include fat (various types and sources), fiber (various types), calories, cholesterol (including HDL/LDL fractions), calcium, alcohol, beta-carotene, and vitamins C and E. Blood samples will be analyzed for beta-carotene and vitamins C and E. Nondietary risk factors of interest include physical activity, hormones, and smoking. Main effects and interactions will be estimated with prevalence odds ratios, using stratified analysis and unconditional logistic regression.